Link prediction on evolving graphs using matrix and tensor factorizations.
Abstract
The data in many disciplines such as social networks, web analysis, etc. is link-based, and the link structure can be exploited for many different data mining tasks. In this paper, we consider the problem of temporal link prediction: Given link data for time periods 1 through T, can we predict the links in time period T + 1? Specifically, we look at bipartite graphs changing over time and consider matrix- and tensor-based methods for predicting links. We present a weight-based method for collapsing multi-year data into a single matrix. We show how the well-known Katz method for link prediction can be extended to bipartite graphs and, moreover, approximated in a scalable way using a truncated singular value decomposition. Using a CANDECOMP/PARAFAC tensor decomposition of the data, we illustrate the usefulness of exploiting the natural three-dimensional structure of temporal link data. Through several numerical experiments, we demonstrate that both matrix- and tensor-based techniques are effective for temporal link prediction despite the inherent difficulty of the problem.
- Authors:
-
- Turkish National Research Institute of Electronics and Cryptology
- Publication Date:
- Research Org.:
- Sandia National Laboratories (SNL), Albuquerque, NM, and Livermore, CA (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1021699
- Report Number(s):
- SAND2010-3823C
TRN: US201117%%291
- DOE Contract Number:
- AC04-94AL85000
- Resource Type:
- Conference
- Resource Relation:
- Conference: Proposed for presentation at the BIT 50 %3CU%2B2013%3E Trends in Numerical Computing held June 17-20, 2011 in Lund, Sweden.
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; FORECASTING; COMPUTER NETWORKS; NETWORK ANALYSIS; DATA ACQUISITION SYSTEMS
Citation Formats
Dunlavy, Daniel M, Acar, Evrim, and Kolda, Tamara Gibson. Link prediction on evolving graphs using matrix and tensor factorizations.. United States: N. p., 2010.
Web.
Dunlavy, Daniel M, Acar, Evrim, & Kolda, Tamara Gibson. Link prediction on evolving graphs using matrix and tensor factorizations.. United States.
Dunlavy, Daniel M, Acar, Evrim, and Kolda, Tamara Gibson. 2010.
"Link prediction on evolving graphs using matrix and tensor factorizations.". United States.
@article{osti_1021699,
title = {Link prediction on evolving graphs using matrix and tensor factorizations.},
author = {Dunlavy, Daniel M and Acar, Evrim and Kolda, Tamara Gibson},
abstractNote = {The data in many disciplines such as social networks, web analysis, etc. is link-based, and the link structure can be exploited for many different data mining tasks. In this paper, we consider the problem of temporal link prediction: Given link data for time periods 1 through T, can we predict the links in time period T + 1? Specifically, we look at bipartite graphs changing over time and consider matrix- and tensor-based methods for predicting links. We present a weight-based method for collapsing multi-year data into a single matrix. We show how the well-known Katz method for link prediction can be extended to bipartite graphs and, moreover, approximated in a scalable way using a truncated singular value decomposition. Using a CANDECOMP/PARAFAC tensor decomposition of the data, we illustrate the usefulness of exploiting the natural three-dimensional structure of temporal link data. Through several numerical experiments, we demonstrate that both matrix- and tensor-based techniques are effective for temporal link prediction despite the inherent difficulty of the problem.},
doi = {},
url = {https://www.osti.gov/biblio/1021699},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue Jun 01 00:00:00 EDT 2010},
month = {Tue Jun 01 00:00:00 EDT 2010}
}